Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method comprising: receiving, by machine logic, a first medical image including a portion of a subject's body, and a lesion having a first size/shape/location with respect to the subject's body; receiving, by machine logic, a second medical image(s) that shows at least a portion of the subject's body, and the lesion having a second size/shape/location with respect to the subject's body; choosing, by machine logic, a plurality of candidate registration points, with each candidate registration point corresponding to a machine logic identifiable location in the subject's body, and with each candidate registration point being shown in both of the first medical image and the second medical image; receiving, by machine logic, a subject medical data set including at least one diagnostic report and at least one structured report based on examination of the subject; applying, by machine logic, a set of lesion-development-related machine logic based rules to determine, based at least in part on the subject medical data set, a potentially compromised area in the second medical image, with the potentially compromised area being an area that is potentially affected by the lesion as shown in the second medical image at the second size/shape/location; selecting, by machine logic, a plurality of selected registration points from the plurality of candidate registration points such that the selected registration points are shown in the second medical image: (i) outside of the potentially compromised area, and (ii) relatively close to the potentially compromised area; and registering, by machine logic, at least the first and second medical images with each other using the set of selected registration points.
Medical imaging systems often struggle to accurately align images of a subject's body over time, especially when lesions or other abnormalities change in size, shape, or location. This misalignment can hinder diagnosis and treatment planning. A method addresses this by improving image registration by accounting for lesion progression. The method receives a first medical image showing a lesion with a first size, shape, and location, and a second medical image showing the lesion with a second size, shape, and location. Machine logic identifies candidate registration points in both images, corresponding to identifiable anatomical locations. The method also receives a subject's medical data, including diagnostic and structured reports. Using lesion-development-related rules, the method determines a potentially compromised area in the second image—an area likely affected by the lesion's progression. From the candidate points, the method selects those outside the compromised area but close to it, ensuring stable registration points. Finally, the method registers the images using these selected points, improving alignment accuracy despite lesion changes. This approach enhances diagnostic reliability by accounting for lesion-induced anatomical changes.
2. The method of claim 1 further comprising: performing machine learning to make a first adjustment to the set of lesion-development-related machine logic rules for determining the potentially compromised area.
This invention relates to medical imaging analysis, specifically detecting and assessing potentially compromised areas in tissue, such as lesions, using machine learning. The problem addressed is the need for accurate and adaptive detection of tissue abnormalities, which can vary in appearance and progression. The method involves analyzing medical imaging data to identify a potentially compromised area based on a set of lesion-development-related machine logic rules. These rules define criteria for detecting and assessing abnormalities in the tissue. The method further includes performing machine learning to refine and adjust these rules, improving the accuracy of identifying compromised areas over time. The machine learning process may involve training models on labeled imaging data, where the labels indicate known lesion characteristics or progression patterns. The adjusted rules are then applied to new imaging data to enhance detection performance. This approach allows the system to adapt to variations in lesion appearance and progression, improving diagnostic reliability. The method may also include preprocessing the imaging data to enhance relevant features and reduce noise, as well as post-processing to refine the detected areas. The overall goal is to provide a robust and adaptive system for early and accurate detection of tissue abnormalities in medical imaging.
3. The method of claim 1 wherein: the lesion is a tumor; the determination of the potentially compromised area includes a prediction of tumor growth.
This invention relates to medical imaging and tumor analysis, specifically predicting tumor growth and identifying potentially compromised areas in tissue. The method involves analyzing medical imaging data to detect a tumor and then determining a region of tissue that may be affected by the tumor's growth. The determination includes predicting how the tumor will expand over time, allowing for early intervention or treatment planning. The system may use machine learning or statistical models trained on historical tumor growth data to make these predictions. The method can be applied to various imaging modalities, such as MRI, CT, or ultrasound, to assess tumor progression in different tissue types. By identifying at-risk areas before they are physically compromised, clinicians can take preventive measures to mitigate further damage. The invention aims to improve early detection and treatment of tumors by leveraging predictive analytics in medical imaging.
4. The method of claim 1 further comprising: receiving a first location data set including information indicative of the first size/shape/location of the lesion; and the determination of the potentially compromised area is further based, at least in part, on the first location data set.
This invention relates to medical imaging and analysis systems for assessing tissue damage, particularly in identifying areas potentially compromised by lesions. The technology addresses the challenge of accurately determining the extent of tissue damage caused by lesions, such as tumors or other abnormalities, by integrating multiple data sources to improve diagnostic precision. The method involves analyzing medical imaging data to detect a lesion and then determining a potentially compromised area of tissue surrounding the lesion. The analysis considers factors such as lesion size, shape, and location to assess the impact on adjacent tissue. Additionally, the method receives a first location data set that provides detailed information about the lesion's size, shape, and location. This data set is used to refine the determination of the potentially compromised area, ensuring a more accurate assessment of tissue damage. By incorporating the lesion's specific characteristics from the location data set, the method enhances the reliability of identifying affected tissue regions, which is critical for treatment planning and monitoring disease progression. The approach improves upon traditional methods by leveraging precise spatial and morphological data to better predict the extent of tissue compromise.
5. The method of claim 1 further comprising: performing organ segmentation of a first organ as the first organ is shown in the first medical image; and performing organ segmentation of the first organ as the first organ is shown in the second medical image.
This invention relates to medical imaging and organ segmentation, addressing the challenge of accurately identifying and delineating organs in multiple medical images. The method involves analyzing at least two medical images, such as CT or MRI scans, to segment a specific organ in each image. Organ segmentation is the process of isolating and defining the boundaries of an organ within the image data. The method ensures that the same organ is consistently identified across different images, which is critical for diagnostic accuracy and treatment planning. By performing segmentation on both images, the method enables comparison or alignment of the organ's position, shape, or condition between the two scans. This can be useful for tracking changes over time, assessing treatment effectiveness, or improving diagnostic confidence. The segmentation process may involve automated algorithms, machine learning techniques, or manual adjustments to ensure precision. The invention enhances medical imaging workflows by providing reliable organ segmentation across multiple images, supporting better clinical decision-making.
6. The method of claim 1 further comprising: masking, by machine logic, the potentially compromised area in the second medical image; and wherein the selection of the selected registration points is based, at least in part, upon the masking.
This invention relates to medical image registration, specifically addressing challenges in aligning images where one image may contain compromised or corrupted regions that could mislead traditional registration algorithms. The problem arises when artifacts, noise, or anatomical changes in a second medical image (e.g., from a scan or imaging modality) distort the image, making accurate alignment with a reference image difficult. Traditional registration methods may incorrectly match these compromised areas, leading to misalignment and diagnostic errors. The solution involves a method that first identifies and masks potentially compromised regions in the second medical image. Machine logic, such as an algorithm or trained model, performs this masking to exclude unreliable areas from the registration process. The method then selects registration points—key landmarks or features used to align the images—based on the masked image. By avoiding compromised regions, the selection of registration points is more accurate, improving the overall alignment between the images. This approach enhances the reliability of medical image registration, particularly in cases where image quality is degraded or anatomical structures are altered. The technique can be applied in various medical imaging contexts, including radiology, surgery, and treatment planning, where precise image alignment is critical.
7. A computer program product (CPP) comprising: a computer readable storage medium configured to store computer code executable by a processor(s) set; and computer code stored on the computer readable storage medium, with the computer code including data and instructions for causing the processor(s) set to perform at least the following operations: receiving, by machine logic, a first medical image including a portion of a subject's body, and a lesion having a first size/shape/location with respect to the subject's body, receiving, by machine logic, a second medical image(s) that shows at least a portion of the subject's body, and the lesion having a second size/shape/location with respect to the subject's body, choosing, by machine logic, a plurality of candidate registration points, with each candidate registration point corresponding to a machine logic identifiable location in the subject's body, and with each candidate registration point being shown in both of the first medical image and the second medical image, receiving, by machine logic, a subject medical data set including at least one diagnostic report and at least one structured report based on examination of the subject, applying, by machine logic, a set of lesion-development-related machine logic based rules to determine, based at least in part on the subject medical data set, a potentially compromised area in the second medical image, with the potentially compromised area being an area that is potentially affected by the lesion as shown in the second medical image at the second size/shape/location, selecting, by machine logic, a plurality of selected registration points from the plurality of candidate registration points such that the selected registration points are shown in the second medical image: (i) outside of the potentially compromised area, and (ii) relatively close to the potentially compromised area, and registering, by machine logic, at least the first and second medical images with each other using the set of selected registration points.
This invention relates to medical imaging and registration techniques for tracking lesions in a subject's body. The problem addressed is accurately aligning multiple medical images of a subject over time, particularly when a lesion has changed in size, shape, or location, which can complicate image registration. The solution involves a computer program product that processes medical images and associated diagnostic data to improve registration accuracy. The system receives a first medical image showing a lesion with a first size, shape, and location, and one or more second medical images showing the lesion at a later time with a second size, shape, and location. The system identifies candidate registration points in both images, corresponding to identifiable anatomical landmarks. It also receives a subject medical data set, including diagnostic and structured reports, to analyze lesion development. Using machine logic-based rules, the system determines a potentially compromised area in the second image—regions likely affected by the lesion's growth or progression. From the candidate points, it selects those outside but near the compromised area to ensure stable registration. Finally, the system registers the images using these selected points, improving alignment accuracy despite lesion changes. This approach enhances diagnostic consistency by minimizing registration errors caused by lesion evolution.
8. The CPP of claim 7 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: performing machine learning to make a first adjustment to the set of lesion-development-related machine logic rules for determining the potentially compromised area.
9. The CPP of claim 7 wherein: the lesion is a tumor; the determination of the potentially compromised area includes a prediction of tumor growth.
This invention relates to medical imaging and tumor analysis, specifically predicting tumor growth and identifying potentially compromised areas in tissue. The system analyzes medical images to detect a tumor and assesses surrounding tissue to predict how the tumor may grow over time. By evaluating factors such as tumor size, location, and surrounding tissue characteristics, the system identifies regions at risk of future tumor expansion. This helps clinicians anticipate treatment needs and plan interventions before the tumor affects critical structures. The prediction model considers biological and anatomical data to estimate growth patterns, improving early intervention strategies. The invention enhances diagnostic accuracy and treatment planning by providing a proactive assessment of tumor progression.
10. The CPP of claim 7 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: receiving a first location data set including information indicative of the first size/shape/location of the lesion; and the determination of the potentially compromised area is further based, at least in part, on the first location data set.
This invention relates to medical imaging and analysis systems for assessing potentially compromised areas in tissue, particularly in the context of lesions. The technology addresses the challenge of accurately identifying and evaluating regions at risk due to lesions, which may be caused by conditions such as tumors, infections, or other pathological changes. The system uses computer code executed by a processor to analyze lesion characteristics and determine the extent of potential compromise in surrounding tissue. The system receives a first location data set containing information about the size, shape, and location of a lesion. This data is used to refine the determination of the potentially compromised area, ensuring that the analysis accounts for the lesion's specific attributes. The processor integrates this information with other relevant data to assess the impact of the lesion on adjacent tissue, improving diagnostic accuracy and treatment planning. The system may also incorporate additional data, such as imaging results or patient history, to enhance the assessment. By leveraging precise lesion data, the invention provides a more reliable method for identifying tissue at risk, supporting better clinical decision-making.
11. The CPP of claim 7 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: performing organ segmentation of a first organ as the first organ is shown in the first medical image; and performing organ segmentation of the first organ as the first organ is shown in the second medical image.
This invention relates to medical imaging and computer-aided diagnosis, specifically improving organ segmentation in multiple medical images. The problem addressed is the difficulty in accurately segmenting organs across different medical images, which is crucial for diagnostic and treatment planning. The solution involves a computer program product (CPP) that processes medical images to segment the same organ in multiple views or modalities. The CPP includes computer code with data and instructions that, when executed by a processor, perform organ segmentation on a first medical image and a second medical image. The segmentation is applied to the same organ in both images, ensuring consistency across different views. This allows for more reliable analysis, comparison, and diagnosis by maintaining accurate organ boundaries in multiple imaging contexts. The segmentation process may involve algorithms that identify and delineate the organ's structure in each image, accounting for variations in imaging techniques or patient positioning. The invention enhances diagnostic accuracy by providing a standardized approach to organ segmentation across different medical images.
12. The CPP of claim 7 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: masking, by machine logic, the potentially compromised area in the second medical image; and wherein the selection of the selected registration points is based, at least in part, upon the masking.
This invention relates to medical imaging and image registration, specifically addressing challenges in accurately aligning medical images when certain areas may be compromised or distorted. The system involves a computer-implemented method for processing medical images, where a first medical image is registered to a second medical image that may contain potentially compromised areas, such as artifacts or distortions. The computer code includes instructions for identifying and masking these compromised regions in the second image to prevent them from interfering with the registration process. The system then selects registration points based on the masked image, ensuring that the alignment is performed using only reliable, uncompromised areas. This approach improves the accuracy of image registration by excluding problematic regions, which is particularly useful in medical applications where precise alignment is critical for diagnosis or treatment planning. The method leverages machine logic to automate the masking process, enhancing efficiency and reducing manual intervention. The overall system ensures robust image registration even in the presence of image artifacts or distortions.
13. A computer system (CS) comprising: a processor(s) set; a computer readable storage medium configured to store computer code executable by a processor(s) set; and computer code stored on the computer readable storage medium, with the computer code including data and instructions for causing the processor(s) set to perform at least the following operations: receiving, by machine logic, a first medical image including a portion of a subject's body, and a lesion having a first size/shape/location with respect to the subject's body, receiving, by machine logic, a second medical image(s) that shows at least a portion of the subject's body, and the lesion having a second size/shape/location with respect to the subject's body, choosing, by machine logic, a plurality of candidate registration points, with each candidate registration point corresponding to a machine logic identifiable location in the subject's body, and with each candidate registration point being shown in both of the first medical image and the second medical image, receiving, by machine logic, a subject medical data set including at least one diagnostic report and at least one structured report based on examination of the subject, applying, by machine logic, a set of lesion-development-related machine logic based rules to determine, based at least in part on the subject medical data set, a potentially compromised area in the second medical image, with the potentially compromised area being an area that is potentially affected by the lesion as shown in the second medical image at the second size/shape/location, selecting, by machine logic, a plurality of selected registration points from the plurality of candidate registration points such that the selected registration points are shown in the second medical image: (i) outside of the potentially compromised area, and (ii) relatively close to the potentially compromised area, and registering, by machine logic, at least the first and second medical images with each other using the set of selected registration points.
A computer system utilizes software to perform medical image registration, particularly for subjects with lesions. The system first receives two medical images of a subject's body, each showing a lesion at different sizes, shapes, and/or locations. It identifies multiple candidate points suitable for image registration that are visible in both images. The system then processes patient medical data, including diagnostic and structured reports, applying a set of rules related to lesion development. Based on this data, it determines a "potentially compromised area" in the second medical image, which represents tissue likely affected by the lesion. From the initial candidates, the system selects specific registration points that are located *outside* this potentially compromised area but are still *relatively close* to it. Finally, the system registers the first and second medical images using only these strategically chosen points. ERROR (embedding): Error: Failed to save embedding: Could not find the 'embedding' column of 'patent_claims' in the schema cache
14. The CS of claim 13 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: performing machine learning to make a first adjustment to the set of lesion-development-related machine logic rules for determining the potentially compromised area.
This invention relates to a computer-implemented system for analyzing medical images to identify potentially compromised areas, such as lesions, in biological tissue. The system addresses the challenge of accurately detecting and assessing lesion development in medical imaging data, which is critical for early diagnosis and treatment planning. The invention leverages machine learning techniques to refine and adjust a set of predefined machine logic rules used to determine potentially compromised areas within the tissue. By dynamically adapting these rules based on learned patterns from the imaging data, the system improves the accuracy and reliability of lesion detection. The machine learning process involves analyzing historical and current imaging data to identify trends, anomalies, and other relevant features that influence lesion development. The adjusted rules are then applied to new imaging data to enhance the precision of identifying areas at risk. This approach reduces false positives and negatives, leading to more effective diagnostic outcomes. The system is designed to integrate with existing medical imaging workflows, providing healthcare professionals with more reliable and actionable insights for patient care.
15. The CS of claim 13 wherein: the lesion is a tumor; the determination of the potentially compromised area includes a prediction of tumor growth.
This invention relates to medical imaging and tumor analysis, specifically predicting tumor growth and identifying potentially compromised areas in tissue. The system uses imaging data to analyze a tumor and surrounding tissue, determining regions at risk of future tumor expansion. The analysis includes predicting how the tumor may grow over time, allowing for early intervention. The system may incorporate machine learning or statistical models trained on historical tumor growth patterns to improve accuracy. The imaging data can be from modalities such as MRI, CT, or ultrasound, and the system may also integrate patient-specific factors like age, genetics, or treatment history to refine predictions. The output includes a visual or numerical assessment of the predicted tumor growth trajectory and the likelihood of tissue compromise, aiding clinicians in treatment planning. The invention aims to improve early detection of high-risk areas, enabling more targeted and timely medical interventions.
16. The CS of claim 13 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: receiving a first location data set including information indicative of the first size/shape/location of the lesion; and the determination of the potentially compromised area is further based, at least in part, on the first location data set.
This invention relates to medical imaging and analysis systems for assessing potentially compromised areas in tissue, particularly in the context of lesions. The system addresses the challenge of accurately identifying and evaluating regions of tissue that may be affected by a lesion, such as in cancer detection or treatment planning. The invention involves a computer-implemented method that processes medical imaging data to determine the extent of a lesion and its potential impact on surrounding tissue. The system receives a first location data set containing information about the size, shape, and location of a lesion. This data is used to analyze the lesion's characteristics and determine a potentially compromised area in the tissue. The analysis considers the lesion's properties to assess how it may affect adjacent or connected regions, improving diagnostic accuracy and treatment planning. The system may also incorporate additional data, such as imaging results or patient-specific information, to refine the assessment. The invention enhances medical decision-making by providing a more precise evaluation of lesion-related risks.
17. The CS of claim 13 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: performing organ segmentation of a first organ as the first organ is shown in the first medical image; and performing organ segmentation of the first organ as the first organ is shown in the second medical image.
This invention relates to medical imaging and computer-aided organ segmentation, addressing the challenge of accurately identifying and isolating organs in multiple medical images. The system uses computer code executed by a processor to analyze medical images, such as those from CT or MRI scans, to segment a specific organ in both a first and a second medical image. The segmentation process involves identifying the boundaries and structure of the organ in each image independently. This allows for comparison or further analysis of the organ's appearance across different images, which can be useful for diagnostic purposes, treatment planning, or monitoring changes over time. The system may also include additional features, such as generating a 3D model of the organ or comparing the segmented organs between the two images to detect differences. The invention improves upon existing methods by automating the segmentation process and ensuring consistency across multiple images, reducing manual effort and potential human error.
18. The CS of claim 13 wherein the computer code further includes data and instructions for causing the processor(s) set to perform at least the following operations: masking, by machine logic, the potentially compromised area in the second medical image; and wherein the selection of the selected registration points is based, at least in part, upon the masking.
This invention relates to medical imaging systems that register or align multiple medical images of the same patient to improve diagnostic accuracy. The problem addressed is the challenge of accurately registering medical images when one or more images may contain compromised areas, such as artifacts, distortions, or regions affected by disease or injury, which can mislead traditional registration algorithms. The system includes a computer-implemented method for processing medical images, where a first medical image is registered to a second medical image that may contain a potentially compromised area. The method involves selecting registration points from the second medical image, where these points are used to align the images. To improve accuracy, the system masks the potentially compromised area in the second medical image before selecting the registration points. The masking step ensures that the compromised region does not influence the selection of registration points, reducing errors in the alignment process. The computer code executing this method includes instructions for performing the masking operation and for selecting registration points based on the masked image. This approach enhances the reliability of medical image registration, particularly in cases where one image contains artifacts or distortions that could otherwise distort the alignment.
Unknown
March 17, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.